Skip to main content
Search
Search
Advanced Search
Skip main navigation
Open access
Short Report
29 July 2026

Mortality risk estimation for autistic older adults: comparing novel machine-learning derived weights versus standard weights for the Charlson Comorbidity Index

Authors: Madison Blake https://orcid.org/0009-0008-8874-9476, Melica Nikahd, J Madison Hyer, Brian W Patterson, Bethany J Wolf, Lauren Bishop, and Brittany N Hand [email protected]Author Info & Affiliations
Volume 15, Number 8
https://doi.org/10.57264/cer-2026-0053
PDFFormats

Abstract

Aim: We aimed to compare Quan and colleagues (2011) established weights for the Charlson Comorbidity Index (CCI) conditions to autism-specific weights for predicting mortality risk in autistic older adults. Materials & methods: We used inpatient healthcare claims from autistic older adults (aged 65+; n = 2829) using the Medicare Standard Analytic Files from 2021 to 2023. We used a machine learning technique called stochastic hill climbing to assign weights to the 12 CCI conditions to maximize predictive ability for 30-day and 1-year mortality. We then compared the resulting area under the curve (AUC) against the established weights. Results: The established weights had poor predictive ability for 30-day (AUC: 0.68; 95% CI: 0.62–0.74) and 1-year mortality (AUC: 0.67; 95% CI: 0.63–0.72). The autism-specific weights also had poor predictive ability for 30-day (AUC: 0.67; 95% CI: 0.61–0.73) and 1-year mortality (AUC: 0.67; 95% CI: 0.62–0.71). Conclusion: The established and autism-specific CCI weights performed similarly in predicting mortality among autistic older adults. Findings may suggest adjusting CCI weights alone is insufficient to accurately predict mortality risk in autistic older adults, and additional health conditions not currently captured by the CCI may need to be added to better predict mortality in this population. Future studies on developing an autism-specific mortality risk index are warranted.

Plain language summary

What is this article about?

Researchers often use tools like the Charlson Comorbidity Index (CCI) to estimate a person's risk of death. The CCI estimates risk of death based on whether they have 12 health conditions like heart failure, lung disease, dementia, cancer and others. Each of these health conditions is assigned a ‘weight’ for how much it increases a person’s risk of death, based on data from the general population. But autistic older adults tend to have different patterns of health conditions than the general population. Therefore, the usual CCI may not accurately estimate risk of death for autistic older adults. We created new weights for the 12 health conditions in the CCI using data from autistic older adults. We expected that these new autism-specific weights would predict risk of death within 30-days and 1-year more accurately for autistic older adults.

What were the results?

The usual CCI and the new autism-specific version we made poorly predicted risk of death within 30-days and 1-year in autistic older adults. Both versions performed about the same.

What do the results of the study mean?

These findings suggest that changing the weights for the 12 health conditions in the CCI did not help better predict risk of death in autistic older adults. Important health factors that influence risk of death in this group may not be included in the CCI at all. Future research should focus on developing a new tool that includes additional relevant health conditions to better predict risk of death in autistic older adults.

Supplementary Material

File (supplementary data.docx)

References

Papers of special note have been highlighted as: • of interest
1.
Dietz PM, Rose CE, McArthur D, Maenner M. National and State estimates of adults with autism spectrum disorder. J. Autism Dev Disord. 50, 4258–4266 (2020).
PubMed
Google Scholar
2.
Grosvenor LP, Croen LA, Lynch FL et al. Autism diagnosis among US Children and Adults, 2011–2022. JAMA Netw. Open 7, e2442218 (2024).
PubMed
Google Scholar
3.
Forsyth L, McSorley M, Rydzewska E. All-cause and cause-specific mortality in people with autism spectrum disorder: a systematic review. Res. Autism Spectr. Disord. 105, 102165 (2023).
Google Scholar
4.
Krantz M, Dalmacy D, Bishop L, Hyer JM, Hand BN. Mortality rate and age of death among Medicare-enrolled autistic older adults. Res. Autism Spectr. Disord. 100, 102077 (2023).
PubMed
Google Scholar
• Highlights that autistic older adults have higher mortality rates compared with nonautistic counterparts.
5.
Croen LA, Zerbo O, Qian Y et al. The health status of adults on the autism spectrum. Autism Int. J. Res. Pract. 19, 814–823 (2015).
PubMed
Google Scholar
6.
Hand BN, Angell AM, Harris L, Carpenter LA. Prevalence of physical and mental health conditions in Medicare-enrolled, autistic older adults. Autism 24, 755–764 (2024).
Google Scholar
• Highlights the physical and mental co-occurring conditions that may disproportionately impact autistic older adults compared with nonautistic peers.
7.
Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J. Chronic Dis. 40, 373–383 (1987).
PubMed
Google Scholar
• This is the original Charlson Comorbidity Index (CCI) that was created to classify 1-year mortality risk based on weighted co-occurring conditions.
8.
Quan H, Sundararajan V, Halfon P et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med. Care 43, 1130–1139 (2005).
PubMed
Google Scholar
9.
Quan H, Li B, Couris CM et al. Updating and validating the Charlson Comorbidity Index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am. J. Epidemiol. 173, 676–682 (2011).
PubMed
Google Scholar
• This study updates and validates new weights for the CCI. This is the most updated and most used version of the CCI in administrative claims studies.
10.
Charlson ME, Carrozzino D, Guidi J, Patierno C. Charlson Comorbidity index: a critical review of clinimetric properties. Psychother. Psychosom. 91, 8–35 (2022).
PubMed
Google Scholar
• Evaluates the clinical usefulness of the CCI and shows that it is useful in providing a valid assessment of a patient’s clinical situation and in determining major diagnostic and prognostic differences among various groups of patients.
11.
Dhakal P, Shostrom V, Al-Kadhimi ZS, Maness LJ, Gundabolu K, Bhatt VR. Usefulness of Charlson Comorbidity Index to predict early mortality and overall survival in older patients with acute myeloid leukemia. Clin. Lymphoma Myeloma Leuk. 20, 804–812 (2020).
PubMed
Google Scholar
12.
Newhouse D, Rangavajla G, Dhande M et al. Abstract 15930: using Charlson Comorbidity Index to predict 1-year all-cause mortality in ICD recipients. Circulation 148, A15930 (2023).
Google Scholar
13.
Zhang N, Lin Q, Jiang H, Zhu H. Age-adjusted Charlson Comorbidity Index as effective predictor for in-hospital mortality of patients with cardiac arrest: a retrospective study. BMC Emerg. Med. 23, 7 (2023).
PubMed
Google Scholar
14.
Gilmore D, Harris L, Longo A, Hand BN. Health status of Medicare-enrolled autistic older adults with and without co-occurring intellectual disability: an analysis of inpatient and institutional outpatient medical claims. Autism Int. J. Res. Pract. 25, 266–274 (2021).
PubMed
Google Scholar
15.
Glasheen WP, Cordier T, Gumpina R, Haugh G, Davis J, Renda A. Charlson Comorbidity Index: ICD-9 update and ICD-10 translation. Am. Health Drug Benefits 12, 188–197 (2019).
PubMed
Google Scholar
16.
U.S. Department of Agriculture Economic Research Service. Rural-urban continuum codes. Economic Research Service. (2023). https://www.ers.usda.gov/data-products/rural-urban-continuum-codes
Google Scholar
17.
Russell SJ, Norvig P. Artificial Intelligence: A Modern Approach. 4th Edition. ISBN-13: 978-0134610993, Pearson, (2021).
Google Scholar
• This is a book that describes hill climbing and stochastic machine learning methods.
18.
Lee SJ, Lindquist K, Segal MR, Covinsky KE. Development and validation of a prognostic index for 4-year mortality in older adults. JAMA 295, 801–808 (2006).
PubMed
Google Scholar
19.
Porock D, Parker-Oliver D, Petroski GF, Rantz M. The MDS mortality risk index: the evolution of a method for predicting 6-month mortality in nursing home residents. BMC Res. Notes 3, 200 (2010).
PubMed
Google Scholar
20.
Schonberg MA, Davis RB, McCarthy EP, Marcantonio ER. Index to predict 5-year mortality of community-dwelling adults aged 65 and older using data from the National Health Interview Survey. J. Gen. Intern. Med. 24, 1115–1122 (2009).
PubMed
Google Scholar
21.
American Geriatrics Society Expert Panel on the Care of Older Adults with Multimorbidity. Guiding principles for the care of older adults with multimorbidity. J. Am. Geriatr. Soc. 60, E1–E25 (2012).
PubMed
Google Scholar
22.
Ferrell BR, Twaddle ML, Melnick A, Meier DE. National Consensus Project clinical practice guidelines for quality palliative care, 4th Edition. J. Palliat. Med. 21, 1684–1689 (2018).
PubMed
Google Scholar
23.
Yourman LC, Lee SJ, Schonberg MA, Widera EW, Smith AK. Prognostic indices for older adults: a systematic review. JAMA 307, 182–192 (2012).
PubMed
Google Scholar
24.
Liu X, Sun X, Sun C et al. Prevalence of epilepsy in autism spectrum disorders: a systematic review and meta-analysis. Autism Int. J. Res. Pract. 26, 33–50 (2022).
PubMed
Google Scholar
25.
Masi A, DeMayo MM, Glozier N, Guastella AJ. An overview of autism spectrum disorder, heterogeneity and treatment options. Neurosci. Bull. 33, 183–193 (2017).
PubMed
Google Scholar
26.
Kunnath AJ, Sack DE, Wilkins CH. Relative predictive value of sociodemographic factors for chronic diseases among All of Us participants: a descriptive analysis. BMC Public Health 24, 405 (2024).
PubMed
Google Scholar
27.
Zhuang Y, Zhao X, Tang S et al. Independent associations of social determinants of health with mortality and added predictive value beyond life's essential 8. J. Health Popul. Nutr. 44, 286 (2025).
PubMed
Google Scholar
28.
Iezzoni LI. Assessing quality using administrative data. Ann. Intern. Med. 127, 666–674 (1997).
PubMed
Google Scholar
29.
Johnson EK, Nelson CP. Utility and pitfalls in the use of administrative databases for outcomes assessment. J. Urol. 190, 17–18 (2013).
PubMed
Google Scholar
30.
Parikh RB, Ferrell WJ, Girard A et al. The impact of machine learning mortality risk prediction on clinician prognostic accuracy and decision support: a randomized vignette study. Med. Decis. Making 45, 690–702 (2025).
PubMed
Google Scholar
• This study underscores that machine learning-based assessments can improve prognostic accuracy and impact clinician decision making.
31.
Riley RD, Collins GS, Kirton L et al. Uncertainty of risk estimates from clinical prediction models: rationale, challenges, and approaches. Br. Med. J. 388, e080749 (2025).
PubMed
Google Scholar
|